PostScript
CORRESPONDENCE
Preserving an integrated view of informatics Biomedical informatics (BMI) covers a broad spectrum of subfields that range in scope from molecules to populations and range in application from research to operations. Subfields include translational bioinformatics (TBI), clinical research informatics (CRI), public health informatics, and others. Recognizing the diversity of BMI, AMIA developed focused meetings for selected subfields including the TBI and CRI ‘Joint Summits,’ in which presentations that address common topics are featured in one overlapping day. The original intent was to complement the general AMIA Annual Symposium and to allow subfields to develop as specializations of BMI. Recently, however, there has been some concern that some subfields may be becoming isolated from the larger medical informatics community to the detriment of the field as a whole. These specialized annual conferences, with their concentration of domain-specific content, together with finite time and travel budgets, may cause a decrease in the content of subfields at the Annual Symposium, further decreasing its relevance and appeal to specialized communities. As specialized subfields mature to the point that they have critical mass to hold their own respective events, there is an increased risk that cross-fertilization and collaboration will be lost. A key question in biomedicine in general and BMI in particular is how best to balance the domain-specific needs of subfields while preserving an integrated approach to the broader discipline. The issue for AMIA in particular is how to create forums for the specialized subfields of informatics in a way that does not balkanize the entire field and cut off communication among the subfields. One might argue that disciplines in biomedicine develop naturally and should be allowed to self-organize in whatever configuration best serves their practitioners. As an example, is it a problem that dermatologists and thoracic surgeons have separate societies and go to separate meetings? Or that computer scientists specialized in human–computer interaction and cryptography go to separate meetings? Probably not. These specialties each have their own specific problems to solve. However, we assert that there are important biomedical e178
problems that can more effectively be, or in some cases can only be, addressed by an integrated BMI approach rather than by separate subfields of BMI. Recent years have seen a trend in healthcare toward personalized (also called precision or individualized) medicine, which refers to ‘coupling established clinical– pathological indexes with state-of-the-art molecular profiling to create diagnostic, prognostic, and therapeutic strategies precisely tailored to each patient’s requirements.’1 Clearly, the realization of personalized medicine will require solving multiple information problems. These include, but are not limited to: ▸ Evidence generation through integrative analysis of novel, high-throughput molecular assays with clinical and population data ▸ Standardization, sharing, and integration of large datasets ▸ Adapting established disease categories to reflect rapidly evolving understanding of the molecular basis of disease ▸ Integrating clinical decision support based on patients’ molecular profiles into operational electronic health records, including visualization of complex genetic and statistical information2 ▸ Development of a ‘learning healthcare system’3 wherein all different types of data collected in the course of patient care are used to generate new knowledge, which then informs future care. Are these clinical informatics problems or bioinformatics problems, research issues or operational challenges? The boundaries begin to blur. What is clear is that separation between fields or subfields is problematic when it impedes progress. Multiple important informatics problems require the combined expertise of public health, clinical, translational, and biomedical informaticians working on academic and applied problems. Multiple organizations contribute to the advancement of personalized medicine. Clinical specialty organizations (eg, American Society of Clinical Oncology)4 appropriately see the delivery of personalized care to be the role of their members. Similarly, scientific and information technology organizations (eg, the Healthcare Information and Management Systems Society) have activities in this area. What then can AMIA and BMI offer in this crowded field? BMI can address the entire spectrum of information problems posed by personalized medicine. Those best positioned to address such problems will have a solid understanding of general informatics (eg, standards and vocabularies; problem
solving, computational, and research methods; decision analysis; interactive application design, data mining; etc) with additional knowledge in specialized subfields such as genetics, clinical practice, and public health. These scientists and practitioners will be able to communicate with each other, as well as with the many non-informatics disciplines that are appropriately involved in personalized medicine. What role will BMI, and by extension AMIA, play in the future of biomedicine? How can members of the informatics community best contribute to the advancement of biomedical research, healthcare, and public health, and how can we best complement our colleagues in clinical practice, information technology, and biological research? These questions are both timely and actionable. Governments and funding agencies are recognizing the power of data to transform businesses and even entire industries, to change the outcomes of elections,5 and to address important problems in healthcare and biomedicine.6 AMIA and the broader science community should work to advance an integrated view of BMI so that the information problems in biomedicine can be addressed as effectively as possible. AMIA should: ▸ Use its programs to promote awareness of the value of an integrated view of informatics. While subfield-specific offerings have their place, AMIA should use its activities to highlight projects that span the biological (molecules↔populations) and application (theory↔practice) spectra. This can be accomplished via thoughtful structuring of scientific meetings and Calls for Participation and diversification of articles in AMIA’s journal. ▸ Increase efforts to engage organizations whose members do not necessarily see themselves as informaticians, but whose work is often within or closely aligned with informatics subfields. Some examples of such organizations include the International Society for Computational Biology, the American Society for Human Genetics, the Association for Computing Machinery, and Institute of Electrical and Electronics Engineers. Successful engagement will require that members of these ‘non-informatics’ organizations see significant value in engaging with AMIA. ▸ Work with academic BMI programs (eg, via the Academic Forum) to develop training curricula that emphasize fundamental informatics methods that are applicable across domains. BMI graduates would thus be well J Am Med Inform Assoc February 2014 Vol 21 No e1
PostScript positioned to leverage methods across domains. ▸ Create communities of individuals whose work crosses the subfields of informatics. An example that is already in place is a community for individuals who work on informatics initiatives under the Clinical and Translational Science Awards (CTSAs), including both applications and methodological innovation. ▸ Work with funding agencies to craft programs and requests for applications that clearly recognize the benefits of an integrated informatics approach. Examples of such programs include the recent NIH ‘Big Data to Knowledge’ (BD2K) program5 and the CTSA program. These programs greatly expanded awareness of BMI among academic medical center faculty and administration, and strongly encouraged multidisciplinary collaboration involving biomedical informaticians. In summary, there are important biomedical problems that require tight integration of expertise that spans the spectrum from the molecules to populations. Similarly, there is synergy between research and practice. An integrated
J Am Med Inform Assoc February 2014 Vol 21 No e1
AMIA reflecting a community of biomedical informaticians with specialized expertise built on a foundation of general information science has a unique niche and a critical role in the future of biomedicine. Elmer V Bernstam,1,2 Jessica D Tenenbaum,3 Gilad J Kuperman4 1
School of Biomedical Informatics, The University of Texas Health Science Center at Houston, Houston, Texas, USA 2 Division of General Internal Medicine, Department of Internal Medicine, Medical School, The University of Texas Health Science Center at Houston, Houston, Texas, USA 3 Duke Translational Medicine Institute, Duke University, Durham, North Carolina, USA 4 New York—Presbyterian Hospital, New York, New York, USA Correspondence to Dr Elmer Bernstam, School of Biomedical Informatics, The University of Texas Health Science Center at Houston, 7000 Fannin St, Suite 600, Houston, TX 77030, USA; [email protected] Contributors All authors listed in the paper provided substantial contribution to conception and design, drafting the article or revising it critically for important intellectual content, and final approval of the completed manuscript. Funding This work was supported in part by NIH NCATS grant number UL1 TR000371 and NIH NCATS grant number UL1 TR000436.
Competing interests None. Provenance and peer review Not commissioned; externally peer reviewed. To cite Bernstam EV, Tenenbaum JD, Kuperman GJ. J Am Med Inform Assoc 2014;21:e178–e179. Received 8 August 2013 Accepted 12 November 2013 Published Online First 28 November 2013 J Am Med Inform Assoc 2014;21:e178–e179. doi:10.1136/amiajnl-2013-002262
REFERENCES 1
2
3
4
5
6
Mirnezami R, Nicholson J, Darzi A. Preparing for precision medicine. N Engl J Med 2012;366: 489–91. Masys DR, Jarvik GP, Abernethy NF, et al. Technical desiderata for the integration of genomic data into Electronic Health Records. J Biomed Inform 2012;45:419–22. Olsen L, Aisner D, McGinnis JM. eds The learning healthcare system: workshop summary (IOM roundtable on evidence-based medicine). The National Academies Press, 2007. Personalized Medicine, Pathway Exploration Common Themes Throughout Program. American Society of Clinical Oncology Daily News, 31 May 2013. http:// am.asco.org/personalized-medicine-pathwayexploration-common-themes-throughout-program Scherer M. How Obama’s data crunchers helped him win. CNN, 8 Nov 2013. http://www.cnn.com/2012/11/ 07/tech/web/obama-campaign-tech-team Big data to knowledge. NIH, 5 Aug 2013. https:// commonfund.nih.gov/bd2k/index.aspx
e179
CORRESPONDENCE
Preserving an integrated view of informatics Biomedical informatics (BMI) covers a broad spectrum of subfields that range in scope from molecules to populations and range in application from research to operations. Subfields include translational bioinformatics (TBI), clinical research informatics (CRI), public health informatics, and others. Recognizing the diversity of BMI, AMIA developed focused meetings for selected subfields including the TBI and CRI ‘Joint Summits,’ in which presentations that address common topics are featured in one overlapping day. The original intent was to complement the general AMIA Annual Symposium and to allow subfields to develop as specializations of BMI. Recently, however, there has been some concern that some subfields may be becoming isolated from the larger medical informatics community to the detriment of the field as a whole. These specialized annual conferences, with their concentration of domain-specific content, together with finite time and travel budgets, may cause a decrease in the content of subfields at the Annual Symposium, further decreasing its relevance and appeal to specialized communities. As specialized subfields mature to the point that they have critical mass to hold their own respective events, there is an increased risk that cross-fertilization and collaboration will be lost. A key question in biomedicine in general and BMI in particular is how best to balance the domain-specific needs of subfields while preserving an integrated approach to the broader discipline. The issue for AMIA in particular is how to create forums for the specialized subfields of informatics in a way that does not balkanize the entire field and cut off communication among the subfields. One might argue that disciplines in biomedicine develop naturally and should be allowed to self-organize in whatever configuration best serves their practitioners. As an example, is it a problem that dermatologists and thoracic surgeons have separate societies and go to separate meetings? Or that computer scientists specialized in human–computer interaction and cryptography go to separate meetings? Probably not. These specialties each have their own specific problems to solve. However, we assert that there are important biomedical e178
problems that can more effectively be, or in some cases can only be, addressed by an integrated BMI approach rather than by separate subfields of BMI. Recent years have seen a trend in healthcare toward personalized (also called precision or individualized) medicine, which refers to ‘coupling established clinical– pathological indexes with state-of-the-art molecular profiling to create diagnostic, prognostic, and therapeutic strategies precisely tailored to each patient’s requirements.’1 Clearly, the realization of personalized medicine will require solving multiple information problems. These include, but are not limited to: ▸ Evidence generation through integrative analysis of novel, high-throughput molecular assays with clinical and population data ▸ Standardization, sharing, and integration of large datasets ▸ Adapting established disease categories to reflect rapidly evolving understanding of the molecular basis of disease ▸ Integrating clinical decision support based on patients’ molecular profiles into operational electronic health records, including visualization of complex genetic and statistical information2 ▸ Development of a ‘learning healthcare system’3 wherein all different types of data collected in the course of patient care are used to generate new knowledge, which then informs future care. Are these clinical informatics problems or bioinformatics problems, research issues or operational challenges? The boundaries begin to blur. What is clear is that separation between fields or subfields is problematic when it impedes progress. Multiple important informatics problems require the combined expertise of public health, clinical, translational, and biomedical informaticians working on academic and applied problems. Multiple organizations contribute to the advancement of personalized medicine. Clinical specialty organizations (eg, American Society of Clinical Oncology)4 appropriately see the delivery of personalized care to be the role of their members. Similarly, scientific and information technology organizations (eg, the Healthcare Information and Management Systems Society) have activities in this area. What then can AMIA and BMI offer in this crowded field? BMI can address the entire spectrum of information problems posed by personalized medicine. Those best positioned to address such problems will have a solid understanding of general informatics (eg, standards and vocabularies; problem
solving, computational, and research methods; decision analysis; interactive application design, data mining; etc) with additional knowledge in specialized subfields such as genetics, clinical practice, and public health. These scientists and practitioners will be able to communicate with each other, as well as with the many non-informatics disciplines that are appropriately involved in personalized medicine. What role will BMI, and by extension AMIA, play in the future of biomedicine? How can members of the informatics community best contribute to the advancement of biomedical research, healthcare, and public health, and how can we best complement our colleagues in clinical practice, information technology, and biological research? These questions are both timely and actionable. Governments and funding agencies are recognizing the power of data to transform businesses and even entire industries, to change the outcomes of elections,5 and to address important problems in healthcare and biomedicine.6 AMIA and the broader science community should work to advance an integrated view of BMI so that the information problems in biomedicine can be addressed as effectively as possible. AMIA should: ▸ Use its programs to promote awareness of the value of an integrated view of informatics. While subfield-specific offerings have their place, AMIA should use its activities to highlight projects that span the biological (molecules↔populations) and application (theory↔practice) spectra. This can be accomplished via thoughtful structuring of scientific meetings and Calls for Participation and diversification of articles in AMIA’s journal. ▸ Increase efforts to engage organizations whose members do not necessarily see themselves as informaticians, but whose work is often within or closely aligned with informatics subfields. Some examples of such organizations include the International Society for Computational Biology, the American Society for Human Genetics, the Association for Computing Machinery, and Institute of Electrical and Electronics Engineers. Successful engagement will require that members of these ‘non-informatics’ organizations see significant value in engaging with AMIA. ▸ Work with academic BMI programs (eg, via the Academic Forum) to develop training curricula that emphasize fundamental informatics methods that are applicable across domains. BMI graduates would thus be well J Am Med Inform Assoc February 2014 Vol 21 No e1
PostScript positioned to leverage methods across domains. ▸ Create communities of individuals whose work crosses the subfields of informatics. An example that is already in place is a community for individuals who work on informatics initiatives under the Clinical and Translational Science Awards (CTSAs), including both applications and methodological innovation. ▸ Work with funding agencies to craft programs and requests for applications that clearly recognize the benefits of an integrated informatics approach. Examples of such programs include the recent NIH ‘Big Data to Knowledge’ (BD2K) program5 and the CTSA program. These programs greatly expanded awareness of BMI among academic medical center faculty and administration, and strongly encouraged multidisciplinary collaboration involving biomedical informaticians. In summary, there are important biomedical problems that require tight integration of expertise that spans the spectrum from the molecules to populations. Similarly, there is synergy between research and practice. An integrated
J Am Med Inform Assoc February 2014 Vol 21 No e1
AMIA reflecting a community of biomedical informaticians with specialized expertise built on a foundation of general information science has a unique niche and a critical role in the future of biomedicine. Elmer V Bernstam,1,2 Jessica D Tenenbaum,3 Gilad J Kuperman4 1
School of Biomedical Informatics, The University of Texas Health Science Center at Houston, Houston, Texas, USA 2 Division of General Internal Medicine, Department of Internal Medicine, Medical School, The University of Texas Health Science Center at Houston, Houston, Texas, USA 3 Duke Translational Medicine Institute, Duke University, Durham, North Carolina, USA 4 New York—Presbyterian Hospital, New York, New York, USA Correspondence to Dr Elmer Bernstam, School of Biomedical Informatics, The University of Texas Health Science Center at Houston, 7000 Fannin St, Suite 600, Houston, TX 77030, USA; [email protected] Contributors All authors listed in the paper provided substantial contribution to conception and design, drafting the article or revising it critically for important intellectual content, and final approval of the completed manuscript. Funding This work was supported in part by NIH NCATS grant number UL1 TR000371 and NIH NCATS grant number UL1 TR000436.
Competing interests None. Provenance and peer review Not commissioned; externally peer reviewed. To cite Bernstam EV, Tenenbaum JD, Kuperman GJ. J Am Med Inform Assoc 2014;21:e178–e179. Received 8 August 2013 Accepted 12 November 2013 Published Online First 28 November 2013 J Am Med Inform Assoc 2014;21:e178–e179. doi:10.1136/amiajnl-2013-002262
REFERENCES 1
2
3
4
5
6
Mirnezami R, Nicholson J, Darzi A. Preparing for precision medicine. N Engl J Med 2012;366: 489–91. Masys DR, Jarvik GP, Abernethy NF, et al. Technical desiderata for the integration of genomic data into Electronic Health Records. J Biomed Inform 2012;45:419–22. Olsen L, Aisner D, McGinnis JM. eds The learning healthcare system: workshop summary (IOM roundtable on evidence-based medicine). The National Academies Press, 2007. Personalized Medicine, Pathway Exploration Common Themes Throughout Program. American Society of Clinical Oncology Daily News, 31 May 2013. http:// am.asco.org/personalized-medicine-pathwayexploration-common-themes-throughout-program Scherer M. How Obama’s data crunchers helped him win. CNN, 8 Nov 2013. http://www.cnn.com/2012/11/ 07/tech/web/obama-campaign-tech-team Big data to knowledge. NIH, 5 Aug 2013. https:// commonfund.nih.gov/bd2k/index.aspx
e179
